Hi Stew,

Thanks for reading and for your comments. I haven’t always felt this way, and still am not totally sold on making it a regular thing for the reasons I (and others) have mentioned. It does, however, seem like a natural fit for showing students that learning is an iterative process and for using assessment for its intended purpose of measuring what students know/don’t know, understand/don’t understand, or can do/can’t do.

You might look at a previous post of mine about my homework policies – I tend to collect all of it as an additional source of information on what students are thinking. These quizzes are just another form of it. It is really easy to get an idea of what mistakes/misconceptions students have from seeing their written work, and it’s even easier when the students are in front of me. This is another push for why I like using quiz situations like this to have the conversation I described in this post. There’s something more real about facing a problem on a quiz rather than homework, so I think I get a more realistic idea of what students can do in a quiz situation. That said, there might be measurement error that comes into play if I do this sort of thing too often and as the line between homework as feedback and quiz as feedback blurs.

I don’t know – I’m still clearly in the experimental stages of figuring this out.

I always appreciate your comments, John.

I think I fit into the same boat as you – I am always answering student questions with more questions since that conversation is always really rich from a learning standpoint. I have almost exclusively been a ‘no-questions’ guy during test and quizzes, in fact I tell any new group of students before the first test I give that I tend to be a jerk when it comes to answering questions during the exam. If there is a legitimate question about formatting, instructions, etc, I will of course answer and inform the rest of the class. When I find them asking questions about content during a unit test, I pretty much repeat “I can’t answer that right now” and walk away.

As for the binary measures for SBG, I think if your standards are defined narrowly in the way I defined them in the rubric, it’s possible. But I’m pretty early on in the SBG learning curve, and I am guessing that if I define everything too narrowly so that skills are tested in complete independence of other skills, I’m going to end up with a list of skills that numbers in the hundreds by the end of the year. That’s something I need to really work on for sure. Whether to go binary or a 1-4 scale is something I’ve thought a lot about, and I’m leaning toward the latter. This is just because I don’t know a consistent way to locate when my assessment switches from “don’t get it” to “get it” and I think I need to define that well if I’m going to go that way and make it fair for students. I’d love to get your input on how you make that distinction.

Finally, I completely agree on conversations – there is a lot you can get out of students talking about their thinking. The biggest issue is time. I can know what students are thinking after a 30 second conversation and have a good idea how to adjust my teaching, but it’s never good enough to give a numerical measure of that understanding.

Thanks for the comment, Shannon.

I agree that giving them the condition of Fn = 0 does kind of take a big chunk of the reasoning out of the problem. In the rubric I used, this concept was really 20% of the points allotted to the problem, so a student that can do the rest is really showing things that other questions will likely be able to assess.

The other piece is that since the student is right in front of me, I CAN look at his/her free body diagram and decide if the Fn = 0 is even a place to start the conversation. If there was another issue, say a tension drawn or an upwards force at that point, this is showing more fundamental errors in the student’s reasoning related to drawing a reasonable FBD for the problem. That’s the nice freedom I have being able to react on the spot to what the students are doing. It also is something I couldn’t do with 25 students in the room, which I am thankful not to have to manage.

The first was how you you connect the rubric with standards through the binary statements and how the rubric guides the conversation with the student about his/her knowledge gaps.

By offering limited assistance on problems, you’re taking to the next level the idea that students should be able to take formative assessments as many times as they like.

Finally, I think you’re right on about how allowing students to retake formative assessments will help change their perceptions about their real purpose.

I’d love to learn more about how you use the data you collect from these assessments to drive your instruction. How are you analyzing it? How do you plan around the data collect?

I’m also not quite sure how I would incorporate this into a binary SBG system I use at the moment. Almost any help from me would show you haven’t mastered it, so that wouldn’t really encourage students to ask me questions.

What I’d really like to do is simply be able to make assessments just conversations between students and myself to figure out where they’re struggling and then have some easy way to report a measure of student understanding. Whenever I do this, I find myself helping way too much, students eventually getting to “a right answer” and me not at all confident that that understanding belongs to the student.

It’s a really thorny issue for me, and I appreciate this post giving me a needed different perspective.